For effective adoptive immunotherapy of cancer, tumor-specific T lymphocytes must be obtained from tumor-bearing patients, activated in vitro and expanded to large numbers, but tumor-induced suppressor cells and the unavailability of tumor antigen may critically limit the expansion of tumor-reactive T cells. In murine tumor models, we have found that T cells from tumor-draining lymph nodes (DLN) are more readily expanded than spleen cells from the same hosts, and that these cells have therapeutic activity in vivo. Low dose cyclophosphamide in vivo and an analogue (mafosfamide) used in vitro can inhibit the ability of suppressor cells to limit cytotoxic T lymphocyte (CTL) growth, without adversely affecting CTL activation. CTL from syngeneic murine P815 tumor-bearing host (TBH) spleens, which would otherwise not grow, could be expanded after in vitro treatment with mafosfamide. Using pharmacologic manipulation of the pathways of T cell signal transduction, via activation of protein kinase C and increasing intracellular Ca++, in place of antigen we could expand T cells which were cytotoxic in vitro and therapeutic in vivo. We have also derived tumor-specific CTL clones from TBH mice which not only depend upon both antigen and exogenous interleukin 2 (IL-2) for growth, but also can attain a "resting" state similar to memory CTL from TBHS. From this resting state, we can distinctly separate the antigen-dependent and IL-2- dependent stages of T cell activation. Thus, we have a unique opportunity to examine in detail the regulation of T cell activation in a tumor- specific, non-transformed and regulable CTL clone. These cloned CTL also have therapeutic activity in vivo and will serve as a model on which to base rational manipulations of lymphocytes from TBH which will maximize T cell growth and preserve anti-tumor activity. Moreover, the above methods of suppressor cell inhibition and pharmacologic T cell activation are applicable to DLN from human patients with breast cancer, but the responsiveness of cells from individual nodes is highly variable. This project aims to: (1) Determine how different methods of T cell activation can be used to expand CTL clones and to delineate the detailed cellular and molecular mechanisms by which regulatory cytokines (IL-2, IL-4 and TGF-Beta) affect the growth of cloned CTL; (2) Determine whether similar methods, combined with inhibition of Ts, function by mafosfamide, can be applied to the expansion of bulk cultures of T cells obtained directly from TBH mice, so that the yield of cells is maximized and anti- tumor activity is maintained; (3) Test the T cells obtained by different, strategies of expansion for therapeutic efficacy in vivo; and (4) Determine how to identify, isolate and expand T cells with potential anti- tumor activity from the DLN of human breast cancer patients, using similar strategies, including mafosfamide treatment and pharmacologic activation of T cells. The proposed studies should lead to a clearer. understanding of tumor-specific CTL activation, regulation and growth and to more rational and more effective strategies for clinical adoptive immunotherapy of cancer.